[OSM-talk] Attribution guideline update

[Christoph Hormann]

On Thursday 20 February 2020, Simon Poole wrote:
> Am 19.02.2020 um 14:24 schrieb Christoph Hormann:
> > In this case the statement that "small maps or multiple data
> > sources" are the only cases where the document does not require
> > visible attribution is wrong.  For example it is later stated that
> > visible attribution is not required if "there is legal or safety or
> > privacy information that needs to be presented with similar or
> > greater prominence to attribution" - which at least in the EU is
> > always the case!
>
> So you agree with us that this is an actual external restraint that
> needs to be considered, and it is not the LWG succumbing to the
> interests of big $$$?

No, the ODbL does not care about outside constraints - if you want to 
use OSM data in a form that does not allow providing proper attribution 
and complying with legal requirements at the same time then you may not 
use OSM data at all.

The formulation i cited is an explicit permission to data users to use 
external legal contraints as an excuse not to attribute visibly.  It 
does not require the legal contraint to even substantially prevent 
proper attribution, its mere existence is declared a valid excuse to 
forego visible attribution.

Anyway - the argument you cited was not actually about this weakening of 
the attribution requirement itself, it was about the fact that the 
guideline draft is inconsistent about this - first claiming 
that "Except for small maps or multiple data sources, as described 
below, attribution must be visible" but then declaring other exceptions 
from the visible attribution requirement.

-- 
Christoph Hormann
http://www.imagico.de/